7a79b20bcf
Model: fpadovani/eng-latn-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed10 Source: Original Platform
2046 lines
54 KiB
JSON
2046 lines
54 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.0811249323958896,
|
|
"eval_steps": 1000,
|
|
"global_step": 1000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 4.816491746902466,
|
|
"epoch": 0.005408328826392645,
|
|
"grad_norm": 12.0,
|
|
"learning_rate": 2e-06,
|
|
"loss": 14.3891,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 10948.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 4.827733516693115,
|
|
"epoch": 0.01081665765278529,
|
|
"grad_norm": 13.375,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 14.3509,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 21455.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 4.862413930892944,
|
|
"epoch": 0.016224986479177934,
|
|
"grad_norm": 17.875,
|
|
"learning_rate": 7e-06,
|
|
"loss": 14.1901,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 32377.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 5.008710432052612,
|
|
"epoch": 0.02163331530557058,
|
|
"grad_norm": 29.5,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 13.7041,
|
|
"mean_token_accuracy": 6.706907879561186e-05,
|
|
"num_tokens": 42276.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 5.984945297241211,
|
|
"epoch": 0.02704164413196322,
|
|
"grad_norm": 44.0,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 12.6083,
|
|
"mean_token_accuracy": 0.00013905229279771448,
|
|
"num_tokens": 55049.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 9.738362789154053,
|
|
"epoch": 0.03244997295835587,
|
|
"grad_norm": 4.125,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 11.0101,
|
|
"mean_token_accuracy": 9.293680195696652e-05,
|
|
"num_tokens": 66792.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.689375400543213,
|
|
"epoch": 0.03785830178474851,
|
|
"grad_norm": 3.53125,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.64,
|
|
"mean_token_accuracy": 0.004824416560586542,
|
|
"num_tokens": 78208.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.698863887786866,
|
|
"epoch": 0.04326663061114116,
|
|
"grad_norm": 2.84375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.2961,
|
|
"mean_token_accuracy": 0.023805590718984602,
|
|
"num_tokens": 88871.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.674648761749268,
|
|
"epoch": 0.048674959437533805,
|
|
"grad_norm": 2.65625,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.9755,
|
|
"mean_token_accuracy": 0.03232901841402054,
|
|
"num_tokens": 100118.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.597569847106934,
|
|
"epoch": 0.05408328826392644,
|
|
"grad_norm": 2.328125,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.7307,
|
|
"mean_token_accuracy": 0.03992431238293648,
|
|
"num_tokens": 113120.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.540758800506591,
|
|
"epoch": 0.05949161709031909,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.5288,
|
|
"mean_token_accuracy": 0.03650712165981531,
|
|
"num_tokens": 123220.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.527715396881103,
|
|
"epoch": 0.06489994591671173,
|
|
"grad_norm": 1.9921875,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.5023,
|
|
"mean_token_accuracy": 0.03586216233670712,
|
|
"num_tokens": 134479.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.531739139556885,
|
|
"epoch": 0.07030827474310439,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.3556,
|
|
"mean_token_accuracy": 0.04139025770127773,
|
|
"num_tokens": 145189.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.52338228225708,
|
|
"epoch": 0.07571660356949703,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.2769,
|
|
"mean_token_accuracy": 0.04220384769141674,
|
|
"num_tokens": 157354.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.495023918151855,
|
|
"epoch": 0.08112493239588967,
|
|
"grad_norm": 1.875,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.2309,
|
|
"mean_token_accuracy": 0.03997939024120569,
|
|
"num_tokens": 168821.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.466780185699463,
|
|
"epoch": 0.08653326122228232,
|
|
"grad_norm": 1.90625,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.1055,
|
|
"mean_token_accuracy": 0.03739558029919863,
|
|
"num_tokens": 180554.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.440871238708496,
|
|
"epoch": 0.09194159004867496,
|
|
"grad_norm": 2.171875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.0271,
|
|
"mean_token_accuracy": 0.03742563612759113,
|
|
"num_tokens": 191229.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.418688297271729,
|
|
"epoch": 0.09734991887506761,
|
|
"grad_norm": 1.9453125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 8.9282,
|
|
"mean_token_accuracy": 0.03555689416825771,
|
|
"num_tokens": 201903.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.375628566741943,
|
|
"epoch": 0.10275824770146025,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.8778,
|
|
"mean_token_accuracy": 0.03980072475969791,
|
|
"num_tokens": 214255.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.296640205383301,
|
|
"epoch": 0.10816657652785289,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.6959,
|
|
"mean_token_accuracy": 0.052101677656173705,
|
|
"num_tokens": 225423.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.19414758682251,
|
|
"epoch": 0.11357490535424554,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.594,
|
|
"mean_token_accuracy": 0.05231698267161846,
|
|
"num_tokens": 237801.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.054014301300048,
|
|
"epoch": 0.11898323418063818,
|
|
"grad_norm": 1.71875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.3897,
|
|
"mean_token_accuracy": 0.051315633580088615,
|
|
"num_tokens": 249195.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 9.872079658508301,
|
|
"epoch": 0.12439156300703083,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.2393,
|
|
"mean_token_accuracy": 0.0522298626601696,
|
|
"num_tokens": 260129.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.810044193267823,
|
|
"epoch": 0.12979989183342347,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.2772,
|
|
"mean_token_accuracy": 0.05217186138033867,
|
|
"num_tokens": 271220.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.666457653045654,
|
|
"epoch": 0.1352082206598161,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.0237,
|
|
"mean_token_accuracy": 0.051312217861413954,
|
|
"num_tokens": 281560.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.364743328094482,
|
|
"epoch": 0.14061654948620878,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.8811,
|
|
"mean_token_accuracy": 0.053314308449625966,
|
|
"num_tokens": 292518.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.2109206199646,
|
|
"epoch": 0.1460248783126014,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.8517,
|
|
"mean_token_accuracy": 0.049220229312777516,
|
|
"num_tokens": 304865.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 8.980883979797364,
|
|
"epoch": 0.15143320713899405,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.6909,
|
|
"mean_token_accuracy": 0.055328205600380896,
|
|
"num_tokens": 315951.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 8.628364181518554,
|
|
"epoch": 0.1568415359653867,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.6116,
|
|
"mean_token_accuracy": 0.05774536021053791,
|
|
"num_tokens": 327094.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.470148944854737,
|
|
"epoch": 0.16224986479177933,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.5585,
|
|
"mean_token_accuracy": 0.058901453390717506,
|
|
"num_tokens": 337801.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.233556652069092,
|
|
"epoch": 0.167658193618172,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.5032,
|
|
"mean_token_accuracy": 0.061500222980976106,
|
|
"num_tokens": 349039.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.088182020187379,
|
|
"epoch": 0.17306652244456464,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.4874,
|
|
"mean_token_accuracy": 0.0693302620202303,
|
|
"num_tokens": 360639.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.017281103134156,
|
|
"epoch": 0.17847485127095727,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.5015,
|
|
"mean_token_accuracy": 0.06765063926577568,
|
|
"num_tokens": 371005.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 7.958820867538452,
|
|
"epoch": 0.1838831800973499,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.3887,
|
|
"mean_token_accuracy": 0.07020511478185654,
|
|
"num_tokens": 383340.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 7.844907283782959,
|
|
"epoch": 0.18929150892374255,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.3382,
|
|
"mean_token_accuracy": 0.07330376878380776,
|
|
"num_tokens": 394682.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 7.825045919418335,
|
|
"epoch": 0.19469983775013522,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.336,
|
|
"mean_token_accuracy": 0.07912327982485294,
|
|
"num_tokens": 405713.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 7.846845293045044,
|
|
"epoch": 0.20010816657652786,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.3066,
|
|
"mean_token_accuracy": 0.07909615188837052,
|
|
"num_tokens": 416208.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.744113254547119,
|
|
"epoch": 0.2055164954029205,
|
|
"grad_norm": 1.65625,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.2175,
|
|
"mean_token_accuracy": 0.0757235523313284,
|
|
"num_tokens": 427595.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.717983436584473,
|
|
"epoch": 0.21092482422931313,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.3271,
|
|
"mean_token_accuracy": 0.08375168889760971,
|
|
"num_tokens": 438273.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.786208295822144,
|
|
"epoch": 0.21633315305570577,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.2509,
|
|
"mean_token_accuracy": 0.07852847650647163,
|
|
"num_tokens": 448948.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.759875535964966,
|
|
"epoch": 0.22174148188209844,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.2839,
|
|
"mean_token_accuracy": 0.07927897274494171,
|
|
"num_tokens": 460838.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.683747482299805,
|
|
"epoch": 0.22714981070849108,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.2887,
|
|
"mean_token_accuracy": 0.082744500041008,
|
|
"num_tokens": 473555.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.688952970504761,
|
|
"epoch": 0.23255813953488372,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.2241,
|
|
"mean_token_accuracy": 0.08608492240309715,
|
|
"num_tokens": 485221.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.682677507400513,
|
|
"epoch": 0.23796646836127636,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.1214,
|
|
"mean_token_accuracy": 0.08501881062984466,
|
|
"num_tokens": 496552.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.638469791412353,
|
|
"epoch": 0.24337479718766902,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.1808,
|
|
"mean_token_accuracy": 0.08574967458844185,
|
|
"num_tokens": 509760.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.626894855499268,
|
|
"epoch": 0.24878312601406166,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.0246,
|
|
"mean_token_accuracy": 0.09106166735291481,
|
|
"num_tokens": 521175.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.549870872497559,
|
|
"epoch": 0.2541914548404543,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.1344,
|
|
"mean_token_accuracy": 0.08684690818190574,
|
|
"num_tokens": 532597.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.58236255645752,
|
|
"epoch": 0.25959978366684694,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.1753,
|
|
"mean_token_accuracy": 0.09069491475820542,
|
|
"num_tokens": 545661.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.5936479568481445,
|
|
"epoch": 0.2650081124932396,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.1579,
|
|
"mean_token_accuracy": 0.0852174036204815,
|
|
"num_tokens": 557336.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.536095857620239,
|
|
"epoch": 0.2704164413196322,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.0178,
|
|
"mean_token_accuracy": 0.0952389270067215,
|
|
"num_tokens": 567398.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.4575268745422365,
|
|
"epoch": 0.27582477014602486,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.0896,
|
|
"mean_token_accuracy": 0.08974430859088897,
|
|
"num_tokens": 579543.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.449991941452026,
|
|
"epoch": 0.28123309897241755,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 6.9161,
|
|
"mean_token_accuracy": 0.09924268275499344,
|
|
"num_tokens": 589218.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.458067178726196,
|
|
"epoch": 0.2866414277988102,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000132,
|
|
"loss": 6.9748,
|
|
"mean_token_accuracy": 0.09499456658959389,
|
|
"num_tokens": 600327.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.406368970870972,
|
|
"epoch": 0.2920497566252028,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 6.926,
|
|
"mean_token_accuracy": 0.09738428071141243,
|
|
"num_tokens": 610923.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.486721420288086,
|
|
"epoch": 0.29745808545159547,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.0091,
|
|
"mean_token_accuracy": 0.09363943114876747,
|
|
"num_tokens": 621138.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.314512109756469,
|
|
"epoch": 0.3028664142779881,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 6.948,
|
|
"mean_token_accuracy": 0.10305705517530442,
|
|
"num_tokens": 631440.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.410723161697388,
|
|
"epoch": 0.30827474310438074,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.9745,
|
|
"mean_token_accuracy": 0.10130783841013909,
|
|
"num_tokens": 642328.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.340037202835083,
|
|
"epoch": 0.3136830719307734,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.9318,
|
|
"mean_token_accuracy": 0.10102855190634727,
|
|
"num_tokens": 652875.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.346016311645508,
|
|
"epoch": 0.319091400757166,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 6.8677,
|
|
"mean_token_accuracy": 0.10212785899639129,
|
|
"num_tokens": 663726.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.447050857543945,
|
|
"epoch": 0.32449972958355866,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.0625,
|
|
"mean_token_accuracy": 0.09114478453993798,
|
|
"num_tokens": 675638.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.26244854927063,
|
|
"epoch": 0.3299080584099513,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.9195,
|
|
"mean_token_accuracy": 0.1022642120718956,
|
|
"num_tokens": 687239.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.387060499191284,
|
|
"epoch": 0.335316387236344,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.9658,
|
|
"mean_token_accuracy": 0.09624839648604393,
|
|
"num_tokens": 698905.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.278276062011718,
|
|
"epoch": 0.34072471606273663,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.9415,
|
|
"mean_token_accuracy": 0.10140556320548058,
|
|
"num_tokens": 710602.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.322119331359863,
|
|
"epoch": 0.34613304488912927,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.9135,
|
|
"mean_token_accuracy": 0.10094785168766976,
|
|
"num_tokens": 721326.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.384740495681763,
|
|
"epoch": 0.3515413737155219,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.894,
|
|
"mean_token_accuracy": 0.10391761958599091,
|
|
"num_tokens": 732972.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.19610185623169,
|
|
"epoch": 0.35694970254191455,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.8612,
|
|
"mean_token_accuracy": 0.10241201519966125,
|
|
"num_tokens": 743624.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.34399447441101,
|
|
"epoch": 0.3623580313683072,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.9483,
|
|
"mean_token_accuracy": 0.10304523780941963,
|
|
"num_tokens": 756384.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.241913795471191,
|
|
"epoch": 0.3677663601946998,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.9174,
|
|
"mean_token_accuracy": 0.10442400574684144,
|
|
"num_tokens": 767256.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.342946195602417,
|
|
"epoch": 0.37317468902109246,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.8438,
|
|
"mean_token_accuracy": 0.10971475169062614,
|
|
"num_tokens": 779059.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.177052688598633,
|
|
"epoch": 0.3785830178474851,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.9283,
|
|
"mean_token_accuracy": 0.11065090894699096,
|
|
"num_tokens": 790291.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.257277774810791,
|
|
"epoch": 0.3839913466738778,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.8328,
|
|
"mean_token_accuracy": 0.10423929244279861,
|
|
"num_tokens": 801450.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.145007610321045,
|
|
"epoch": 0.38939967550027044,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.7503,
|
|
"mean_token_accuracy": 0.11073209047317505,
|
|
"num_tokens": 812428.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.148734092712402,
|
|
"epoch": 0.3948080043266631,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.8508,
|
|
"mean_token_accuracy": 0.10246649086475372,
|
|
"num_tokens": 823644.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.288735342025757,
|
|
"epoch": 0.4002163331530557,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.8351,
|
|
"mean_token_accuracy": 0.11553217023611069,
|
|
"num_tokens": 834064.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.1784649848937985,
|
|
"epoch": 0.40562466197944835,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.7847,
|
|
"mean_token_accuracy": 0.10547300353646279,
|
|
"num_tokens": 845207.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.121596479415894,
|
|
"epoch": 0.411032990805841,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.843,
|
|
"mean_token_accuracy": 0.10698517560958862,
|
|
"num_tokens": 855564.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.189912986755371,
|
|
"epoch": 0.41644131963223363,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.8547,
|
|
"mean_token_accuracy": 0.11624484211206436,
|
|
"num_tokens": 866959.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.079056596755981,
|
|
"epoch": 0.42184964845862627,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.7793,
|
|
"mean_token_accuracy": 0.11480710729956627,
|
|
"num_tokens": 877319.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.181234741210938,
|
|
"epoch": 0.4272579772850189,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.866,
|
|
"mean_token_accuracy": 0.10589642673730851,
|
|
"num_tokens": 888037.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.0435555934906,
|
|
"epoch": 0.43266630611141155,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.6004,
|
|
"mean_token_accuracy": 0.1158117413520813,
|
|
"num_tokens": 898775.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.104381990432739,
|
|
"epoch": 0.43807463493780424,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.7622,
|
|
"mean_token_accuracy": 0.11117781847715377,
|
|
"num_tokens": 909513.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.088195323944092,
|
|
"epoch": 0.4434829637641969,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.8126,
|
|
"mean_token_accuracy": 0.10974752902984619,
|
|
"num_tokens": 920730.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.043075609207153,
|
|
"epoch": 0.4488912925905895,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.7279,
|
|
"mean_token_accuracy": 0.11817020252346992,
|
|
"num_tokens": 931457.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.0011475563049315,
|
|
"epoch": 0.45429962141698216,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.7087,
|
|
"mean_token_accuracy": 0.11752742975950241,
|
|
"num_tokens": 942572.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.129669713973999,
|
|
"epoch": 0.4597079502433748,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.7241,
|
|
"mean_token_accuracy": 0.10921066999435425,
|
|
"num_tokens": 953132.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.0359748840332035,
|
|
"epoch": 0.46511627906976744,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.7533,
|
|
"mean_token_accuracy": 0.11189360246062278,
|
|
"num_tokens": 963551.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.023745107650757,
|
|
"epoch": 0.4705246078961601,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.7177,
|
|
"mean_token_accuracy": 0.11238613426685333,
|
|
"num_tokens": 975297.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 6.960337781906128,
|
|
"epoch": 0.4759329367225527,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.7147,
|
|
"mean_token_accuracy": 0.11535659208893775,
|
|
"num_tokens": 986910.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.105753707885742,
|
|
"epoch": 0.48134126554894535,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.8383,
|
|
"mean_token_accuracy": 0.11518610119819642,
|
|
"num_tokens": 997058.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.0269519805908205,
|
|
"epoch": 0.48674959437533805,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.7099,
|
|
"mean_token_accuracy": 0.10921002328395843,
|
|
"num_tokens": 1008261.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.005563545227051,
|
|
"epoch": 0.4921579232017307,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.6246,
|
|
"mean_token_accuracy": 0.11358757168054581,
|
|
"num_tokens": 1019193.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 6.977084970474243,
|
|
"epoch": 0.4975662520281233,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.7447,
|
|
"mean_token_accuracy": 0.1139349453151226,
|
|
"num_tokens": 1031227.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 6.943169641494751,
|
|
"epoch": 0.502974580854516,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.5781,
|
|
"mean_token_accuracy": 0.1172634594142437,
|
|
"num_tokens": 1042963.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 6.869647264480591,
|
|
"epoch": 0.5083829096809086,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.6377,
|
|
"mean_token_accuracy": 0.1132569208741188,
|
|
"num_tokens": 1053709.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.069565773010254,
|
|
"epoch": 0.5137912385073012,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.6675,
|
|
"mean_token_accuracy": 0.12598269656300545,
|
|
"num_tokens": 1064527.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 6.852591276168823,
|
|
"epoch": 0.5191995673336939,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.7166,
|
|
"mean_token_accuracy": 0.11791721880435943,
|
|
"num_tokens": 1075127.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 6.959320735931397,
|
|
"epoch": 0.5246078961600865,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.7096,
|
|
"mean_token_accuracy": 0.11324949637055397,
|
|
"num_tokens": 1086581.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 6.855788326263427,
|
|
"epoch": 0.5300162249864792,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.6395,
|
|
"mean_token_accuracy": 0.12406286075711251,
|
|
"num_tokens": 1097818.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 6.895084428787231,
|
|
"epoch": 0.5354245538128718,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.62,
|
|
"mean_token_accuracy": 0.11416960656642913,
|
|
"num_tokens": 1108043.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 6.967974853515625,
|
|
"epoch": 0.5408328826392644,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.6878,
|
|
"mean_token_accuracy": 0.11347125247120857,
|
|
"num_tokens": 1119875.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 6.858421468734742,
|
|
"epoch": 0.5462412114656571,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.5843,
|
|
"mean_token_accuracy": 0.1187442697584629,
|
|
"num_tokens": 1132406.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.007649278640747,
|
|
"epoch": 0.5516495402920497,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.7718,
|
|
"mean_token_accuracy": 0.1115761086344719,
|
|
"num_tokens": 1143828.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 6.889253616333008,
|
|
"epoch": 0.5570578691184424,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.5729,
|
|
"mean_token_accuracy": 0.11763741374015808,
|
|
"num_tokens": 1155482.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 6.731452751159668,
|
|
"epoch": 0.5624661979448351,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.5244,
|
|
"mean_token_accuracy": 0.12875338792800903,
|
|
"num_tokens": 1166847.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.8908110618591305,
|
|
"epoch": 0.5678745267712277,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.5924,
|
|
"mean_token_accuracy": 0.12007522955536842,
|
|
"num_tokens": 1177419.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.906198453903198,
|
|
"epoch": 0.5732828555976204,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.7698,
|
|
"mean_token_accuracy": 0.11545055732131004,
|
|
"num_tokens": 1188975.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.836439752578736,
|
|
"epoch": 0.578691184424013,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.6162,
|
|
"mean_token_accuracy": 0.11988403424620628,
|
|
"num_tokens": 1200252.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.827834510803223,
|
|
"epoch": 0.5840995132504057,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.657,
|
|
"mean_token_accuracy": 0.12472999319434167,
|
|
"num_tokens": 1211864.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.788261890411377,
|
|
"epoch": 0.5895078420767983,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.524,
|
|
"mean_token_accuracy": 0.12068347483873368,
|
|
"num_tokens": 1222429.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.91283164024353,
|
|
"epoch": 0.5949161709031909,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.632,
|
|
"mean_token_accuracy": 0.11880290359258652,
|
|
"num_tokens": 1233782.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.792628812789917,
|
|
"epoch": 0.6003244997295836,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.5843,
|
|
"mean_token_accuracy": 0.11936102360486985,
|
|
"num_tokens": 1244309.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.7369684219360355,
|
|
"epoch": 0.6057328285559762,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.535,
|
|
"mean_token_accuracy": 0.12587283179163933,
|
|
"num_tokens": 1256403.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.848638105392456,
|
|
"epoch": 0.6111411573823688,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.6314,
|
|
"mean_token_accuracy": 0.11637956202030182,
|
|
"num_tokens": 1268758.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.7654791355133055,
|
|
"epoch": 0.6165494862087615,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.5245,
|
|
"mean_token_accuracy": 0.12454744204878807,
|
|
"num_tokens": 1279355.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.76825737953186,
|
|
"epoch": 0.6219578150351541,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.5023,
|
|
"mean_token_accuracy": 0.11673943027853965,
|
|
"num_tokens": 1290619.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.779633808135986,
|
|
"epoch": 0.6273661438615468,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.4739,
|
|
"mean_token_accuracy": 0.12195988073945045,
|
|
"num_tokens": 1301851.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.81631326675415,
|
|
"epoch": 0.6327744726879394,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.6271,
|
|
"mean_token_accuracy": 0.12264949083328247,
|
|
"num_tokens": 1313260.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.702813005447387,
|
|
"epoch": 0.638182801514332,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.614,
|
|
"mean_token_accuracy": 0.12174173891544342,
|
|
"num_tokens": 1324189.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.848341035842895,
|
|
"epoch": 0.6435911303407247,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.6393,
|
|
"mean_token_accuracy": 0.12240146771073342,
|
|
"num_tokens": 1335765.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.741462898254395,
|
|
"epoch": 0.6489994591671173,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.5711,
|
|
"mean_token_accuracy": 0.12267256677150726,
|
|
"num_tokens": 1347653.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.757457304000854,
|
|
"epoch": 0.65440778799351,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.565,
|
|
"mean_token_accuracy": 0.12561324387788772,
|
|
"num_tokens": 1360201.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.749183225631714,
|
|
"epoch": 0.6598161168199026,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.4399,
|
|
"mean_token_accuracy": 0.12647488713264465,
|
|
"num_tokens": 1370974.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.667228031158447,
|
|
"epoch": 0.6652244456462953,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.5024,
|
|
"mean_token_accuracy": 0.12279477640986443,
|
|
"num_tokens": 1383855.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.6716227531433105,
|
|
"epoch": 0.670632774472688,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.4609,
|
|
"mean_token_accuracy": 0.13709836080670357,
|
|
"num_tokens": 1395041.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.799122095108032,
|
|
"epoch": 0.6760411032990806,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.5701,
|
|
"mean_token_accuracy": 0.12464237436652184,
|
|
"num_tokens": 1406102.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.585179328918457,
|
|
"epoch": 0.6814494321254733,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.3918,
|
|
"mean_token_accuracy": 0.13039295971393586,
|
|
"num_tokens": 1417713.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.634333515167237,
|
|
"epoch": 0.6868577609518659,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.545,
|
|
"mean_token_accuracy": 0.12864828929305078,
|
|
"num_tokens": 1428526.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.689204120635987,
|
|
"epoch": 0.6922660897782585,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.4914,
|
|
"mean_token_accuracy": 0.130169028788805,
|
|
"num_tokens": 1438849.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.685839700698852,
|
|
"epoch": 0.6976744186046512,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.562,
|
|
"mean_token_accuracy": 0.1240805745124817,
|
|
"num_tokens": 1450776.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.58751163482666,
|
|
"epoch": 0.7030827474310438,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.4104,
|
|
"mean_token_accuracy": 0.12494389489293098,
|
|
"num_tokens": 1462552.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.5898017406463625,
|
|
"epoch": 0.7084910762574365,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.4285,
|
|
"mean_token_accuracy": 0.12942259460687638,
|
|
"num_tokens": 1473867.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.534844923019409,
|
|
"epoch": 0.7138994050838291,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.4365,
|
|
"mean_token_accuracy": 0.1297524243593216,
|
|
"num_tokens": 1485233.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.6757666110992435,
|
|
"epoch": 0.7193077339102217,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.5013,
|
|
"mean_token_accuracy": 0.1329445168375969,
|
|
"num_tokens": 1496919.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.615731143951416,
|
|
"epoch": 0.7247160627366144,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.4425,
|
|
"mean_token_accuracy": 0.1267342150211334,
|
|
"num_tokens": 1508695.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.712345457077026,
|
|
"epoch": 0.730124391563007,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.4929,
|
|
"mean_token_accuracy": 0.13088276386260986,
|
|
"num_tokens": 1519629.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.592201662063599,
|
|
"epoch": 0.7355327203893997,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.3946,
|
|
"mean_token_accuracy": 0.13325998485088347,
|
|
"num_tokens": 1530144.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.504761219024658,
|
|
"epoch": 0.7409410492157923,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.3993,
|
|
"mean_token_accuracy": 0.1304351307451725,
|
|
"num_tokens": 1541013.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.65454888343811,
|
|
"epoch": 0.7463493780421849,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.4689,
|
|
"mean_token_accuracy": 0.1292506165802479,
|
|
"num_tokens": 1553344.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.584673690795898,
|
|
"epoch": 0.7517577068685776,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.5258,
|
|
"mean_token_accuracy": 0.12957922369241714,
|
|
"num_tokens": 1565298.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.584612560272217,
|
|
"epoch": 0.7571660356949702,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.4559,
|
|
"mean_token_accuracy": 0.13932728618383408,
|
|
"num_tokens": 1577645.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.577200269699096,
|
|
"epoch": 0.7625743645213628,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.4113,
|
|
"mean_token_accuracy": 0.13460915088653563,
|
|
"num_tokens": 1588387.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.715465974807739,
|
|
"epoch": 0.7679826933477556,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.5222,
|
|
"mean_token_accuracy": 0.1325209140777588,
|
|
"num_tokens": 1599837.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.541527509689331,
|
|
"epoch": 0.7733910221741482,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.4538,
|
|
"mean_token_accuracy": 0.1247769333422184,
|
|
"num_tokens": 1610401.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.497765398025512,
|
|
"epoch": 0.7787993510005409,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.4115,
|
|
"mean_token_accuracy": 0.1288266658782959,
|
|
"num_tokens": 1622688.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.657041215896607,
|
|
"epoch": 0.7842076798269335,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.4753,
|
|
"mean_token_accuracy": 0.12559469044208527,
|
|
"num_tokens": 1633594.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.517681550979614,
|
|
"epoch": 0.7896160086533262,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.3937,
|
|
"mean_token_accuracy": 0.1302385449409485,
|
|
"num_tokens": 1645861.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.380044412612915,
|
|
"epoch": 0.7950243374797188,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.2871,
|
|
"mean_token_accuracy": 0.13864490687847136,
|
|
"num_tokens": 1656903.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.560413932800293,
|
|
"epoch": 0.8004326663061114,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.3941,
|
|
"mean_token_accuracy": 0.13524588868021964,
|
|
"num_tokens": 1668130.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.4031064987182615,
|
|
"epoch": 0.8058409951325041,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.2856,
|
|
"mean_token_accuracy": 0.1413346864283085,
|
|
"num_tokens": 1678228.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.61671633720398,
|
|
"epoch": 0.8112493239588967,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.4299,
|
|
"mean_token_accuracy": 0.1359289936721325,
|
|
"num_tokens": 1690339.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.550224161148071,
|
|
"epoch": 0.8166576527852893,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.4545,
|
|
"mean_token_accuracy": 0.1291230209171772,
|
|
"num_tokens": 1701067.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.591836929321289,
|
|
"epoch": 0.822065981611682,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.3858,
|
|
"mean_token_accuracy": 0.13830180689692498,
|
|
"num_tokens": 1712835.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.482460069656372,
|
|
"epoch": 0.8274743104380746,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.3431,
|
|
"mean_token_accuracy": 0.1339413307607174,
|
|
"num_tokens": 1723432.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.485507011413574,
|
|
"epoch": 0.8328826392644673,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.3816,
|
|
"mean_token_accuracy": 0.13742958456277848,
|
|
"num_tokens": 1734134.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.505664300918579,
|
|
"epoch": 0.8382909680908599,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.3263,
|
|
"mean_token_accuracy": 0.13703825324773788,
|
|
"num_tokens": 1744723.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.503205013275147,
|
|
"epoch": 0.8436992969172525,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.4692,
|
|
"mean_token_accuracy": 0.13435956612229347,
|
|
"num_tokens": 1755269.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.569448661804199,
|
|
"epoch": 0.8491076257436452,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.4695,
|
|
"mean_token_accuracy": 0.13084665387868882,
|
|
"num_tokens": 1766344.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.478793334960938,
|
|
"epoch": 0.8545159545700378,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.4252,
|
|
"mean_token_accuracy": 0.12769865095615388,
|
|
"num_tokens": 1779052.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.41169285774231,
|
|
"epoch": 0.8599242833964305,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.2906,
|
|
"mean_token_accuracy": 0.13171118646860122,
|
|
"num_tokens": 1789754.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.330496978759766,
|
|
"epoch": 0.8653326122228231,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.2156,
|
|
"mean_token_accuracy": 0.13964635729789734,
|
|
"num_tokens": 1801161.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.5482982158660885,
|
|
"epoch": 0.8707409410492158,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.3145,
|
|
"mean_token_accuracy": 0.13513469696044922,
|
|
"num_tokens": 1810892.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.267117166519165,
|
|
"epoch": 0.8761492698756085,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.3747,
|
|
"mean_token_accuracy": 0.1325235478579998,
|
|
"num_tokens": 1821574.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.579942226409912,
|
|
"epoch": 0.8815575987020011,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.4147,
|
|
"mean_token_accuracy": 0.13671720623970032,
|
|
"num_tokens": 1833342.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.4688621997833256,
|
|
"epoch": 0.8869659275283938,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.2885,
|
|
"mean_token_accuracy": 0.13777956888079643,
|
|
"num_tokens": 1843768.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.47753415107727,
|
|
"epoch": 0.8923742563547864,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.3655,
|
|
"mean_token_accuracy": 0.13368769288063048,
|
|
"num_tokens": 1855569.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.481969928741455,
|
|
"epoch": 0.897782585181179,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.3524,
|
|
"mean_token_accuracy": 0.12639456540346145,
|
|
"num_tokens": 1866684.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.346806287765503,
|
|
"epoch": 0.9031909140075717,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.2795,
|
|
"mean_token_accuracy": 0.13589733690023423,
|
|
"num_tokens": 1878700.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.458641433715821,
|
|
"epoch": 0.9085992428339643,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.3092,
|
|
"mean_token_accuracy": 0.13823288977146148,
|
|
"num_tokens": 1889970.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.346112823486328,
|
|
"epoch": 0.914007571660357,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.2211,
|
|
"mean_token_accuracy": 0.13803198635578157,
|
|
"num_tokens": 1902008.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.384577941894531,
|
|
"epoch": 0.9194159004867496,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.451,
|
|
"mean_token_accuracy": 0.13807493597269058,
|
|
"num_tokens": 1913253.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.430484962463379,
|
|
"epoch": 0.9248242293131422,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.2842,
|
|
"mean_token_accuracy": 0.14151261672377585,
|
|
"num_tokens": 1923408.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.387599515914917,
|
|
"epoch": 0.9302325581395349,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.3452,
|
|
"mean_token_accuracy": 0.1367344081401825,
|
|
"num_tokens": 1935315.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.312086009979248,
|
|
"epoch": 0.9356408869659275,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.1907,
|
|
"mean_token_accuracy": 0.13912060707807541,
|
|
"num_tokens": 1946524.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.40140061378479,
|
|
"epoch": 0.9410492157923201,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.3316,
|
|
"mean_token_accuracy": 0.13535718396306037,
|
|
"num_tokens": 1957561.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.32704906463623,
|
|
"epoch": 0.9464575446187128,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.2669,
|
|
"mean_token_accuracy": 0.14048450738191604,
|
|
"num_tokens": 1969089.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.425353097915649,
|
|
"epoch": 0.9518658734451054,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.3031,
|
|
"mean_token_accuracy": 0.13773690536618233,
|
|
"num_tokens": 1980341.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.304427719116211,
|
|
"epoch": 0.9572742022714981,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.2545,
|
|
"mean_token_accuracy": 0.14644995033740998,
|
|
"num_tokens": 1991220.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.370989036560059,
|
|
"epoch": 0.9626825310978907,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.3384,
|
|
"mean_token_accuracy": 0.13986072912812234,
|
|
"num_tokens": 2002650.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.394271993637085,
|
|
"epoch": 0.9680908599242833,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.245,
|
|
"mean_token_accuracy": 0.1418061763048172,
|
|
"num_tokens": 2012884.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.391276216506958,
|
|
"epoch": 0.9734991887506761,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.3143,
|
|
"mean_token_accuracy": 0.1372893489897251,
|
|
"num_tokens": 2023543.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.445915412902832,
|
|
"epoch": 0.9789075175770687,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.377,
|
|
"mean_token_accuracy": 0.13422725573182107,
|
|
"num_tokens": 2034789.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.311644124984741,
|
|
"epoch": 0.9843158464034614,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.2856,
|
|
"mean_token_accuracy": 0.1408335119485855,
|
|
"num_tokens": 2046719.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.227353477478028,
|
|
"epoch": 0.989724175229854,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.2471,
|
|
"mean_token_accuracy": 0.14040217846632003,
|
|
"num_tokens": 2057016.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.333889055252075,
|
|
"epoch": 0.9951325040562466,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.2208,
|
|
"mean_token_accuracy": 0.13608154356479646,
|
|
"num_tokens": 2067898.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.261037720574273,
|
|
"epoch": 1.0,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.2446,
|
|
"mean_token_accuracy": 0.14002363714906904,
|
|
"num_tokens": 2076889.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.367987632751465,
|
|
"epoch": 1.0054083288263926,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.1674,
|
|
"mean_token_accuracy": 0.13393356874585152,
|
|
"num_tokens": 2089228.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.271270132064819,
|
|
"epoch": 1.0108166576527853,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.1582,
|
|
"mean_token_accuracy": 0.14539326280355452,
|
|
"num_tokens": 2101098.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.24772367477417,
|
|
"epoch": 1.016224986479178,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.1042,
|
|
"mean_token_accuracy": 0.15308297127485276,
|
|
"num_tokens": 2113081.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.293310213088989,
|
|
"epoch": 1.0216333153055706,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.0795,
|
|
"mean_token_accuracy": 0.14808320105075837,
|
|
"num_tokens": 2123356.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.367437076568604,
|
|
"epoch": 1.0270416441319632,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.1807,
|
|
"mean_token_accuracy": 0.14733164012432098,
|
|
"num_tokens": 2134388.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.116773366928101,
|
|
"epoch": 1.0324499729583558,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.0548,
|
|
"mean_token_accuracy": 0.14803084284067153,
|
|
"num_tokens": 2145209.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.321672821044922,
|
|
"epoch": 1.0378583017847485,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.1102,
|
|
"mean_token_accuracy": 0.14471006840467454,
|
|
"num_tokens": 2157146.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.179027318954468,
|
|
"epoch": 1.043266630611141,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.1315,
|
|
"mean_token_accuracy": 0.1431720107793808,
|
|
"num_tokens": 2168721.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.207623815536499,
|
|
"epoch": 1.0486749594375337,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.1216,
|
|
"mean_token_accuracy": 0.14734558016061783,
|
|
"num_tokens": 2179213.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.2188718795776365,
|
|
"epoch": 1.0540832882639264,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.1466,
|
|
"mean_token_accuracy": 0.1441408984363079,
|
|
"num_tokens": 2191224.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.3757565975189205,
|
|
"epoch": 1.059491617090319,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.2076,
|
|
"mean_token_accuracy": 0.14605461582541465,
|
|
"num_tokens": 2203665.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.216717863082886,
|
|
"epoch": 1.0648999459167117,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.1271,
|
|
"mean_token_accuracy": 0.14381530210375787,
|
|
"num_tokens": 2215136.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.285787582397461,
|
|
"epoch": 1.0703082747431043,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.1576,
|
|
"mean_token_accuracy": 0.14630142003297805,
|
|
"num_tokens": 2226375.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.071659278869629,
|
|
"epoch": 1.075716603569497,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.0121,
|
|
"mean_token_accuracy": 0.15493061542510986,
|
|
"num_tokens": 2236097.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.3767985820770265,
|
|
"epoch": 1.0811249323958896,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.2663,
|
|
"mean_token_accuracy": 0.14112632349133492,
|
|
"num_tokens": 2247964.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.0811249323958896,
|
|
"eval_entropy": 6.043832895972512,
|
|
"eval_loss": 6.424855709075928,
|
|
"eval_mean_token_accuracy": 0.14066070311448792,
|
|
"eval_num_tokens": 2247964.0,
|
|
"eval_runtime": 2.5746,
|
|
"eval_samples_per_second": 1365.265,
|
|
"eval_steps_per_second": 170.901,
|
|
"step": 1000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 9240,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 1000,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 3510135917568000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|